CoolFace
Modelpublic

Realgon/N_roberta_imdb_padding60model

sourceHugging Facemitupdated 3y agoView on Hugging Face
0likes17downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Nrobertaimdb_padding60model

This model is a fine-tuned version of roberta-base on the imdb dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.4811
  • —Accuracy: 0.9500

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • —learning_rate: 2e-05
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 20

Training results

Training LossEpochStepValidation LossAccuracy
0.21571.015630.20200.9366
0.17162.031260.17570.9467
0.11353.046890.26010.9442
0.08344.062520.24980.9485
0.05335.078150.34800.9452
0.04416.093780.35480.9371
0.03197.0109410.32570.9474
0.02648.0125040.39320.9457
0.02399.0140670.33670.9469
0.018510.0156300.45000.94
0.01811.0171930.38710.9470
0.015312.0187560.42060.9456
0.010113.0203190.40270.9492
0.00514.0218820.47010.9477
0.005115.0234450.44540.9484
0.001316.0250080.50150.9493
0.00717.0265710.50110.9476
0.002218.0281340.47980.95
0.004719.0296970.47940.9498
0.001420.0312600.48110.9500

Framework versions

  • —Transformers 4.33.2
  • —Pytorch 2.0.1+cu117
  • —Datasets 2.14.5
  • —Tokenizers 0.13.3